Older hearing-impaired listeners experience speech recognition deficits in noise that persist even after their audibility is improved using suitable amplification. There is a lack of association between speech recognition in noise and most tasks involving the perception of simple intensity, frequency, and temporal acoustic features. The proposed research is aimed at studying spectral integration in a group of young normal-hearing, older normal-hearing, and older hearing-impaired listeners, in order to investigate age and/or hearing loss-related deficits in across-channel listening. Spectral integration for intensive and temporal features will be measured for non-speech and speech stimuli using experimental conditions that selectively emphasize either feature encoding or feature combination and comparison across channels. Further, experiments on spectral integration of intensive and temporal features will be designed to differentiate deficits in feature combination and feature comparisons. Comparisons across these various tasks will determine if age- and/or hearing loss- related deficits in across-channel processing are a consequence of peripheral encoding or more central processing deficits. Finally, the proposed work will determine if across-channel integration abilities for non-speech and speech stimuli are associated with older and hearing-impaired listeners'speech recognition performance in noise. The results of the proposed work will further our understanding of individual differences in across- channel listening and speech recognition in noise. In the long term, they have implications for improving diagnostic and rehabilitative strategies for older and hearing-impaired listeners. PUBLIC HEALTH RELEVANCE: The proposed research will determine the association between age- and hearing loss-related changes in spectral integration of intensity and temporal information and age- and hearing loss-related deficits in speech identification in noise.